Potential outliers

Removed team 17080712_1 with over 600 collisions and massive distance deviation

Collisions

team age age_drone aus_born aus_born_drone aus_years aus_years_drone dic_use dic_use_drone eng_fl eng_fl_drone gender gender_drone collisions_overall speed_overall time_taken_overall distance_overall distance_overall_deviation events_missed_overall issue issue_note
17032915_2 18 18 1 1 NA NA 1 NA 2 1 2 2 589 5.10237 2462.302 12131.93 531.9305 5 FALSE NA
17040314_1 18 18 1 2 NA 1 NA 1 1 2 1 2 370 10.04094 1263.552 12678.38 1078.3817 0 FALSE NA
17080810_2 28 19 1 1 NA NA NA NA 1 1 2 2 391 11.12127 1260.406 14502.92 2902.9222 0 FALSE NA
## [1] "All scores on psych variables are within 1.5 SD of the mean"

Speed

team age age_drone aus_born aus_born_drone aus_years aus_years_drone dic_use dic_use_drone eng_fl eng_fl_drone gender gender_drone collisions_overall speed_overall time_taken_overall distance_overall distance_overall_deviation events_missed_overall issue issue_note
17040414_2 18 19 1 1 NA NA NA NA 1 1 2 2 228 5.700897 1943.420 11398.96 -201.0416 3 FALSE NA
17040711_1 20 18 1 2 NA 17 NA NA 1 1 2 2 60 4.360415 2650.336 11829.80 229.7985 0 FALSE NA
17080810_2 28 19 1 1 NA NA NA NA 1 1 2 2 391 11.121270 1260.406 14502.92 2902.9222 0 FALSE NA
## [1] "All scores on psych variables are within 1.5 SD of the mean"

Distance deviation

team age age_drone aus_born aus_born_drone aus_years aus_years_drone dic_use dic_use_drone eng_fl eng_fl_drone gender gender_drone collisions_overall speed_overall time_taken_overall distance_overall distance_overall_deviation events_missed_overall issue issue_note
16101114_1 18 18 2 1 12 NA NA NA 1 1 1 2 119 8.843025 1929.837 16970.16 5370.1591 1 FALSE NA
16110214_1 18 19 2 2 13 1 2 3 2 2 1 2 208 7.894563 2110.132 17066.86 5466.8634 2 FALSE NA
17032409_1 25 18 1 1 NA NA 2 NA 2 1 2 2 234 7.476407 2187.731 16259.47 4659.4685 0 FALSE NA
17081512_2 18 19 1 2 NA 5 NA 3 1 2 1 2 291 9.974998 1252.345 12589.61 989.6076 0 TRUE major driver-drone networking error - codriver saw no-go signs in place of all arrows and saw roughly 10% of the driver’s traffic
## [1] "All scores on psych variables are within 1.5 SD of the mean"

Collisions analyses with outliers removed

## [1] "Team 17032915_2 removed from analyses"
## # A tibble: 8 x 2
##   rowname              collisions_overall
##   <chr>                             <dbl>
## 1 confidence_drone                   0.28
## 2 incongruent_errors                 0.28
## 3 inhibitory_cost                    0.27
## 4 prop_female                        0.31
## 5 sit.awareness                     -0.34
## 6 sit.awareness_driver              -0.31
## 7 switch_errors                      0.38
## 8 terrible_codriver                  0.37
##                    vars  n mean   sd median trimmed  mad   min  max range
## confidence_drone      1 52 0.00 1.00   0.00    0.04 0.89 -2.75 1.75  4.50
## incongruent_errors    2 52 0.00 1.00  -0.30   -0.16 0.86 -0.88 3.19  4.07
## inhibitory_cost       3 52 0.00 1.00  -0.12   -0.07 0.79 -2.31 2.87  5.18
## prop_female*          4 52 2.31 0.67   2.00    2.38 1.48  1.00 3.00  2.00
## switch_errors         5 52 0.00 1.00  -0.08   -0.14 0.74 -1.07 3.40  4.47
## terrible_codriver     6 52 0.00 1.00  -0.17   -0.15 0.75 -1.34 3.55  4.89
##                     skew kurtosis   se
## confidence_drone   -0.34    -0.07 0.14
## incongruent_errors  1.29     1.02 0.14
## inhibitory_cost     0.72     1.08 0.14
## prop_female*       -0.43    -0.86 0.09
## switch_errors       1.14     1.28 0.14
## terrible_codriver   1.57     2.73 0.14
## 
## Call:
## lm(formula = tmp$collisions_overall ~ ., data = var_std)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -131.621  -38.377   -1.337   35.032  162.174 
## 
## Coefficients:
##                    Estimate Std. Error t value Pr(>|t|)   
## (Intercept)          102.22      30.05   3.402  0.00144 **
## confidence_drone      23.94      10.68   2.242  0.03005 * 
## incongruent_errors    20.88      11.29   1.848  0.07129 . 
## inhibitory_cost       22.32      10.21   2.187  0.03407 * 
## prop_female0.5        56.75      33.86   1.676  0.10085   
## prop_female1          68.16      34.13   1.997  0.05204 . 
## switch_errors         26.80      11.05   2.426  0.01945 * 
## terrible_codriver     26.33      10.42   2.528  0.01515 * 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 71.75 on 44 degrees of freedom
## Multiple R-squared:  0.486,  Adjusted R-squared:  0.4042 
## F-statistic: 5.942 on 7 and 44 DF,  p-value: 6.695e-05
## 
##                     Correlations                     
## ----------------------------------------------------
## Variable              Zero Order    Partial    Part  
## ----------------------------------------------------
## confidence_drone           0.276      0.320    0.242 
## incongruent_errors         0.280      0.268    0.200 
## inhibitory_cost            0.274      0.313    0.236 
## prop_female0.5            -0.036      0.245    0.181 
## prop_female1               0.227      0.288    0.216 
## switch_errors              0.380      0.343    0.262 
## terrible_codriver          0.370      0.356    0.273 
## ----------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.7269         0
## Farrar Chi-Square:        15.3641         0
## Red Indicator:             0.1397         0
## Sum of Lambda Inverse:     6.7053         0
## Theil's Method:           -1.7594         0
## Condition Number:         12.4171         0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                       VIF    TOL     Wi     Fi Leamer   CVIF Klein
## confidence_drone   1.1086 0.9021 0.9989 1.2758 0.9498 1.4801     0
## incongruent_errors 1.1910 0.8396 1.7576 2.2447 0.9163 1.5902     0
## inhibitory_cost    1.0299 0.9710 0.2750 0.3512 0.9854 1.3750     0
## prop_female        1.0963 0.9122 0.8856 1.1311 0.9551 1.4636     0
## switch_errors      1.2051 0.8298 1.8872 2.4103 0.9109 1.6090     0
## terrible_codriver  1.0744 0.9307 0.6846 0.8743 0.9647 1.4345     0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## prop_female , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.4748 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [1] "Team 17040314_1 removed from analyses"
## # A tibble: 9 x 2
##   rowname               collisions_overall
##   <chr>                              <dbl>
## 1 incongruent_errors                  0.36
## 2 inconsistent_codriver               0.28
## 3 inhibitory_cost                     0.26
## 4 prop_female                         0.37
## 5 repeat_errors                       0.26
## 6 sex_driver                         -0.33
## 7 sit.awareness                      -0.39
## 8 sit.awareness_driver               -0.4 
## 9 terrible_codriver                   0.3 
##                       vars  n mean   sd median trimmed  mad   min  max
## incongruent_errors       1 52 0.00 1.00  -0.31   -0.17 0.83 -0.88 3.06
## inconsistent_codriver    2 52 0.00 1.00  -0.17   -0.08 1.04 -1.36 2.49
## inhibitory_cost          3 52 0.00 1.00  -0.10   -0.07 0.76 -2.32 2.87
## prop_female*             4 52 2.33 0.68   2.00    2.40 1.48  1.00 3.00
## repeat_errors            5 52 0.00 1.00   0.10   -0.12 0.89 -1.10 2.51
## terrible_codriver        6 52 0.00 1.00  -0.16   -0.16 0.73 -1.34 3.59
##                       range  skew kurtosis   se
## incongruent_errors     3.94  1.25     0.71 0.14
## inconsistent_codriver  3.84  0.61    -0.52 0.14
## inhibitory_cost        5.18  0.70     1.07 0.14
## prop_female*           2.00 -0.48    -0.85 0.09
## repeat_errors          3.60  0.84     0.13 0.14
## terrible_codriver      4.93  1.64     3.00 0.14
## 
## Call:
## lm(formula = tmp$collisions_overall ~ ., data = var_std)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -146.384  -42.882   -7.766   37.663  305.942 
## 
## Coefficients:
##                       Estimate Std. Error t value Pr(>|t|)  
## (Intercept)             99.802     37.492   2.662   0.0108 *
## incongruent_errors      34.159     15.359   2.224   0.0313 *
## inconsistent_codriver   13.420     13.631   0.984   0.3303  
## inhibitory_cost         22.092     12.583   1.756   0.0861 .
## prop_female0.5          44.930     42.709   1.052   0.2985  
## prop_female1            94.473     41.969   2.251   0.0294 *
## repeat_errors            7.677     15.057   0.510   0.6127  
## terrible_codriver       22.137     13.470   1.643   0.1074  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 89.27 on 44 degrees of freedom
## Multiple R-squared:  0.3964, Adjusted R-squared:  0.3004 
## F-statistic: 4.128 on 7 and 44 DF,  p-value: 0.001451
## 
##                      Correlations                       
## -------------------------------------------------------
## Variable                 Zero Order    Partial    Part  
## -------------------------------------------------------
## incongruent_errors            0.359      0.318    0.260 
## inconsistent_codriver         0.285      0.147    0.115 
## inhibitory_cost               0.257      0.256    0.206 
## prop_female0.5               -0.145      0.157    0.123 
## prop_female1                  0.319      0.321    0.264 
## repeat_errors                 0.259      0.077    0.060 
## terrible_codriver             0.303      0.240    0.192 
## -------------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.5942         0
## Farrar Chi-Square:        25.0725         1
## Red Indicator:             0.1780         0
## Sum of Lambda Inverse:     7.1863         0
## Theil's Method:           -1.0709         0
## Condition Number:          7.7182         0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                          VIF    TOL     Wi     Fi Leamer   CVIF Klein
## incongruent_errors    1.3875 0.7207 3.5649 4.5530 0.8490 1.9387     0
## inconsistent_codriver 1.1572 0.8641 1.4466 1.8476 0.9296 1.6170     0
## inhibitory_cost       1.0115 0.9886 0.1060 0.1354 0.9943 1.4134     0
## prop_female           1.0563 0.9467 0.5176 0.6610 0.9730 1.4759     0
## repeat_errors         1.4128 0.7078 3.7974 4.8499 0.8413 1.9740     0
## terrible_codriver     1.1610 0.8613 1.4811 1.8917 0.9281 1.6222     0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## inconsistent_codriver , inhibitory_cost , repeat_errors , terrible_codriver , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.3963 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [1] "Team 17080810_2 removed from analyses"
## # A tibble: 7 x 2
##   rowname               collisions_overall
##   <chr>                              <dbl>
## 1 incongruent_errors                 0.43 
## 2 inconsistent_codriver              0.290
## 3 prop_female                        0.31 
## 4 repeat_errors                      0.27 
## 5 sit.awareness                     -0.35 
## 6 sit.awareness_driver              -0.38 
## 7 terrible_codriver                  0.32 
##                       vars  n mean   sd median trimmed  mad   min  max
## incongruent_errors       1 52 0.00 1.00  -0.35   -0.16 0.84 -0.91 3.04
## inconsistent_codriver    2 52 0.00 1.00  -0.17   -0.08 1.04 -1.36 2.48
## prop_female*             3 52 2.31 0.67   2.00    2.38 1.48  1.00 3.00
## repeat_errors            4 52 0.00 1.00   0.10   -0.12 0.89 -1.10 2.51
## terrible_codriver        5 52 0.00 1.00  -0.17   -0.15 0.73 -1.34 3.55
##                       range  skew kurtosis   se
## incongruent_errors     3.95  1.18     0.57 0.14
## inconsistent_codriver  3.84  0.60    -0.54 0.14
## prop_female*           2.00 -0.43    -0.86 0.09
## repeat_errors          3.60  0.84     0.13 0.14
## terrible_codriver      4.89  1.58     2.77 0.14
## 
## Call:
## lm(formula = tmp$collisions_overall ~ ., data = var_std)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -129.67  -42.39  -14.54   37.62  299.49 
## 
## Coefficients:
##                       Estimate Std. Error t value Pr(>|t|)   
## (Intercept)            103.733     37.188   2.789  0.00771 **
## incongruent_errors      41.666     14.903   2.796  0.00759 **
## inconsistent_codriver   14.065     13.514   1.041  0.30353   
## prop_female0.5          47.051     42.316   1.112  0.27208   
## prop_female1            84.166     41.609   2.023  0.04906 * 
## repeat_errors            5.313     14.550   0.365  0.71671   
## terrible_codriver       26.375     13.260   1.989  0.05279 . 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 88.48 on 45 degrees of freedom
## Multiple R-squared:  0.3839, Adjusted R-squared:  0.3017 
## F-statistic: 4.673 on 6 and 45 DF,  p-value: 0.0009001
## 
##                      Correlations                       
## -------------------------------------------------------
## Variable                 Zero Order    Partial    Part  
## -------------------------------------------------------
## incongruent_errors            0.431      0.385    0.327 
## inconsistent_codriver         0.294      0.153    0.122 
## prop_female0.5               -0.065      0.164    0.130 
## prop_female1                  0.242      0.289    0.237 
## repeat_errors                 0.265      0.054    0.043 
## terrible_codriver             0.324      0.284    0.233 
## -------------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.6418         0
## Farrar Chi-Square:        21.2859         1
## Red Indicator:             0.2038         0
## Sum of Lambda Inverse:     5.9820         0
## Theil's Method:           -0.7487         0
## Condition Number:          5.8572         0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                          VIF    TOL     Wi     Fi Leamer   CVIF Klein
## incongruent_errors    1.3135 0.7613 3.6835 5.0158 0.8725 1.7729     0
## inconsistent_codriver 1.1585 0.8632 1.8629 2.5367 0.9291 1.5637     0
## prop_female           1.0383 0.9631 0.4496 0.6122 0.9814 1.4014     0
## repeat_errors         1.3270 0.7536 3.8419 5.2315 0.8681 1.7911     0
## terrible_codriver     1.1447 0.8736 1.7002 2.3152 0.9347 1.5450     0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## inconsistent_codriver , repeat_errors , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.3835 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [[1]]
## NULL
## 
## [[2]]
## NULL
## 
## [[3]]
## NULL

Speed analyses with outliers removed

## [1] "Team 17040414_2 removed from analyses"
## # A tibble: 12 x 2
##    rowname               speed_overall
##    <chr>                         <dbl>
##  1 agreeableness                -0.3  
##  2 helpful_exchange             -0.46 
##  3 inconsistent_codriver        -0.32 
##  4 leadership                   -0.33 
##  5 leadership_co_driver         -0.45 
##  6 neuroticism_drone             0.31 
##  7 prop_female                  -0.47 
##  8 resilience_drone             -0.27 
##  9 sex_driver                    0.570
## 10 sit.awareness                 0.28 
## 11 sit.awareness_driver          0.570
## 12 switch_time_drone             0.3  
##                       vars  n mean   sd median trimmed  mad   min  max
## agreeableness            1 52 0.00 1.00   0.05    0.05 0.87 -2.31 1.61
## helpful_exchange         2 52 0.00 1.00   0.05   -0.03 1.08 -1.82 2.44
## inconsistent_codriver    3 52 0.00 1.00  -0.06   -0.07 1.22 -1.40 2.49
## neuroticism_drone        4 52 0.00 1.00  -0.03   -0.03 1.02 -2.09 2.39
## prop_female*             5 52 2.31 0.67   2.00    2.38 1.48  1.00 3.00
## resilience_drone         6 46 0.00 1.00   0.07    0.00 1.16 -1.86 2.20
## switch_time_drone        7 52 0.00 1.00  -0.26   -0.08 0.94 -1.50 2.36
##                       range  skew kurtosis   se
## agreeableness          3.92 -0.38    -0.19 0.14
## helpful_exchange       4.26  0.21    -0.54 0.14
## inconsistent_codriver  3.89  0.58    -0.50 0.14
## neuroticism_drone      4.48  0.35    -0.53 0.14
## prop_female*           2.00 -0.43    -0.86 0.09
## resilience_drone       4.06  0.06    -0.69 0.15
## switch_time_drone      3.86  0.64    -0.64 0.14
## 
## Call:
## lm(formula = tmp$speed_overall ~ ., data = var_std)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -2.99863 -0.57799  0.06526  0.70773  2.62132 
## 
## Coefficients:
##                       Estimate Std. Error t value Pr(>|t|)    
## (Intercept)            9.18305    0.60697  15.129   <2e-16 ***
## agreeableness         -0.06200    0.21294  -0.291   0.7725    
## helpful_exchange      -0.45523    0.21652  -2.102   0.0424 *  
## inconsistent_codriver -0.24358    0.19376  -1.257   0.2166    
## neuroticism_drone      0.26026    0.21349   1.219   0.2305    
## prop_female0.5        -0.30897    0.66618  -0.464   0.6455    
## prop_female1          -1.34691    0.66638  -2.021   0.0505 .  
## resilience_drone      -0.06134    0.19979  -0.307   0.7605    
## switch_time_drone      0.34781    0.19698   1.766   0.0857 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.2 on 37 degrees of freedom
##   (6 observations deleted due to missingness)
## Multiple R-squared:  0.531,  Adjusted R-squared:  0.4296 
## F-statistic: 5.236 on 8 and 37 DF,  p-value: 0.0002035
## 
##                       Correlations                       
## --------------------------------------------------------
## Variable                 Zero Order    Partial     Part  
## --------------------------------------------------------
## agreeableness                -0.305     -0.048    -0.033 
## helpful_exchange             -0.501     -0.327    -0.237 
## inconsistent_codriver        -0.353     -0.202    -0.142 
## neuroticism_drone             0.343      0.197     0.137 
## prop_female0.5                0.439     -0.076    -0.052 
## prop_female1                 -0.539     -0.315    -0.228 
## resilience_drone             -0.270     -0.050    -0.035 
## switch_time_drone             0.296      0.279     0.199 
## --------------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.3880         0
## Farrar Chi-Square:        40.0307         1
## Red Indicator:             0.2275         0
## Sum of Lambda Inverse:     8.9938         0
## Theil's Method:           -1.6473         0
## Condition Number:         42.7239         1
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                          VIF    TOL     Wi     Fi Leamer     CVIF Klein
## agreeableness         1.4488 0.6902 2.9173 3.5905 0.8308 208.6639     0
## helpful_exchange      1.5373 0.6505 3.4926 4.2986 0.8065 221.4121     0
## inconsistent_codriver 1.1935 0.8379 1.2576 1.5478 0.9154 171.8881     0
## neuroticism_drone     1.3000 0.7692 1.9500 2.4000 0.8771 187.2312     0
## prop_female           1.1318 0.8835 0.8569 1.0547 0.9400 163.0115     0
## resilience_drone      1.2438 0.8040 1.5847 1.9504 0.8967 179.1363     0
## switch_time_drone     1.1385 0.8783 0.9003 1.1081 0.9372 163.9730     0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## agreeableness , inconsistent_codriver , neuroticism_drone , resilience_drone , switch_time_drone , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.5223 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [1] "Team 17040711_1 removed from analyses"
## # A tibble: 11 x 2
##    rowname               speed_overall
##    <chr>                         <dbl>
##  1 agreeableness                -0.35 
##  2 helpful_exchange             -0.42 
##  3 inconsistent_codriver        -0.31 
##  4 leadership                   -0.290
##  5 leadership_co_driver         -0.31 
##  6 prop_female                  -0.47 
##  7 repeat_time_drone             0.31 
##  8 sex_driver                    0.580
##  9 sit.awareness                 0.39 
## 10 sit.awareness_driver          0.570
## 11 switch_time_drone             0.32 
##                       vars  n mean   sd median trimmed  mad   min  max
## agreeableness            1 52 0.00 1.00   0.02    0.06 1.15 -2.30 1.57
## helpful_exchange         2 52 0.00 1.00   0.01   -0.03 1.01 -1.78 2.44
## inconsistent_codriver    3 52 0.00 1.00  -0.05   -0.07 1.23 -1.37 2.47
## prop_female*             4 52 2.31 0.67   2.00    2.38 1.48  1.00 3.00
## repeat_time_drone        5 52 0.00 1.00  -0.25   -0.12 0.90 -1.49 3.23
## switch_time_drone        6 52 0.00 1.00  -0.32   -0.08 0.87 -1.49 2.36
##                       range  skew kurtosis   se
## agreeableness          3.87 -0.40    -0.27 0.14
## helpful_exchange       4.22  0.26    -0.59 0.14
## inconsistent_codriver  3.85  0.58    -0.55 0.14
## prop_female*           2.00 -0.43    -0.86 0.09
## repeat_time_drone      4.71  1.04     0.68 0.14
## switch_time_drone      3.85  0.65    -0.64 0.14
## 
## Call:
## lm(formula = tmp$speed_overall ~ ., data = var_std)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.3104 -0.6196  0.0890  0.7295  3.3269 
## 
## Coefficients:
##                       Estimate Std. Error t value Pr(>|t|)    
## (Intercept)             9.0952     0.4830  18.830   <2e-16 ***
## agreeableness          -0.1800     0.1959  -0.919   0.3633    
## helpful_exchange       -0.3715     0.1907  -1.948   0.0578 .  
## inconsistent_codriver  -0.1775     0.1759  -1.009   0.3184    
## prop_female0.5         -0.2820     0.5592  -0.504   0.6166    
## prop_female1           -1.2830     0.5408  -2.372   0.0221 *  
## repeat_time_drone      -0.1819     0.3527  -0.516   0.6087    
## switch_time_drone       0.5478     0.3354   1.633   0.1095    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.154 on 44 degrees of freedom
## Multiple R-squared:  0.4611, Adjusted R-squared:  0.3753 
## F-statistic: 5.378 on 7 and 44 DF,  p-value: 0.0001682
## 
##                       Correlations                       
## --------------------------------------------------------
## Variable                 Zero Order    Partial     Part  
## --------------------------------------------------------
## agreeableness                -0.353     -0.137    -0.102 
## helpful_exchange             -0.421     -0.282    -0.216 
## inconsistent_codriver        -0.310     -0.150    -0.112 
## prop_female0.5                0.419     -0.076    -0.056 
## prop_female1                 -0.531     -0.337    -0.263 
## repeat_time_drone             0.307     -0.078    -0.057 
## switch_time_drone             0.317      0.239     0.181 
## --------------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.1423         0
## Farrar Chi-Square:        93.9006         1
## Red Indicator:             0.2993         0
## Sum of Lambda Inverse:    13.6096         0
## Theil's Method:            0.0920         0
## Condition Number:         26.8449         0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                          VIF    TOL      Wi      Fi Leamer    CVIF Klein
## agreeableness         1.4711 0.6798  4.3340  5.5353 0.8245  4.3982     0
## helpful_exchange      1.3590 0.7359  3.3025  4.2179 0.8578  4.0630     0
## inconsistent_codriver 1.1782 0.8488  1.6390  2.0933 0.9213  3.5224     0
## prop_female           1.0891 0.9182  0.8193  1.0463 0.9582  3.2560     0
## repeat_time_drone     4.2916 0.2330 30.2827 38.6763 0.4827 12.8307     1
## switch_time_drone     4.2207 0.2369 29.6308 37.8437 0.4867 12.6189     1
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## agreeableness , inconsistent_codriver , repeat_time_drone , switch_time_drone , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.4511 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [1] "Team 17080810_2 removed from analyses"
## # A tibble: 11 x 2
##    rowname               speed_overall
##    <chr>                         <dbl>
##  1 agreeableness                -0.39 
##  2 helpful_exchange             -0.39 
##  3 inconsistent_codriver        -0.28 
##  4 leadership                   -0.290
##  5 leadership_co_driver         -0.35 
##  6 prop_female                  -0.55 
##  7 repeat_time_drone             0.31 
##  8 sex_driver                    0.65 
##  9 sit.awareness                 0.41 
## 10 sit.awareness_driver          0.65 
## 11 switch_time_drone             0.34 
##                       vars  n mean   sd median trimmed  mad   min  max
## agreeableness            1 52 0.00 1.00   0.05    0.05 0.87 -2.31 1.61
## helpful_exchange         2 52 0.00 1.00   0.05   -0.03 1.08 -1.81 2.43
## inconsistent_codriver    3 52 0.00 1.00  -0.17   -0.08 1.04 -1.36 2.48
## prop_female*             4 52 2.31 0.67   2.00    2.38 1.48  1.00 3.00
## repeat_time_drone        5 52 0.00 1.00  -0.27   -0.11 0.90 -1.53 3.23
## switch_time_drone        6 52 0.00 1.00  -0.26   -0.08 0.94 -1.50 2.36
##                       range  skew kurtosis   se
## agreeableness          3.92 -0.38    -0.19 0.14
## helpful_exchange       4.24  0.21    -0.56 0.14
## inconsistent_codriver  3.84  0.60    -0.54 0.14
## prop_female*           2.00 -0.43    -0.86 0.09
## repeat_time_drone      4.76  1.01     0.68 0.14
## switch_time_drone      3.86  0.64    -0.64 0.14
## 
## Call:
## lm(formula = tmp$speed_overall ~ ., data = var_std)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.8337 -0.6767  0.1897  0.7843  1.8078 
## 
## Coefficients:
##                       Estimate Std. Error t value Pr(>|t|)    
## (Intercept)             9.0702     0.4677  19.395  < 2e-16 ***
## agreeableness          -0.3170     0.1907  -1.663  0.10350    
## helpful_exchange       -0.2320     0.1883  -1.232  0.22440    
## inconsistent_codriver  -0.1236     0.1697  -0.728  0.47045    
## prop_female0.5         -0.2331     0.5395  -0.432  0.66778    
## prop_female1           -1.5847     0.5242  -3.023  0.00416 ** 
## repeat_time_drone      -0.2637     0.3325  -0.793  0.43198    
## switch_time_drone       0.6297     0.3194   1.972  0.05495 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.115 on 44 degrees of freedom
## Multiple R-squared:  0.5327, Adjusted R-squared:  0.4584 
## F-statistic: 7.167 on 7 and 44 DF,  p-value: 1.013e-05
## 
##                       Correlations                       
## --------------------------------------------------------
## Variable                 Zero Order    Partial     Part  
## --------------------------------------------------------
## agreeableness                -0.390     -0.243    -0.171 
## helpful_exchange             -0.394     -0.183    -0.127 
## inconsistent_codriver        -0.281     -0.109    -0.075 
## prop_female0.5                0.484     -0.065    -0.045 
## prop_female1                 -0.613     -0.415    -0.312 
## repeat_time_drone             0.315     -0.119    -0.082 
## switch_time_drone             0.337      0.285     0.203 
## --------------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.1422         0
## Farrar Chi-Square:        93.9611         1
## Red Indicator:             0.3007         0
## Sum of Lambda Inverse:    13.4487         0
## Theil's Method:           -0.1784         0
## Condition Number:         26.8407         0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                          VIF    TOL      Wi      Fi Leamer    CVIF Klein
## agreeableness         1.4926 0.6700  4.5319  5.7881 0.8185  7.1622     0
## helpful_exchange      1.4143 0.7071  3.8118  4.8683 0.8409  6.7866     0
## inconsistent_codriver 1.1768 0.8497  1.6268  2.0777 0.9218  5.6469     0
## prop_female           1.0884 0.9187  0.8137  1.0393 0.9585  5.2229     0
## repeat_time_drone     4.1511 0.2409 28.9899 37.0251 0.4908 19.9187     1
## switch_time_drone     4.1254 0.2424 28.7539 36.7238 0.4923 19.7957     1
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## agreeableness , helpful_exchange , inconsistent_codriver , repeat_time_drone , switch_time_drone , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.5099 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [[1]]
## NULL
## 
## [[2]]
## NULL
## 
## [[3]]
## NULL

Distance deviation analyses with outliers removed

## [1] "Team 16101114_1 removed from analyses"
## # A tibble: 7 x 2
##   rowname               distance_overall_deviation
##   <chr>                                      <dbl>
## 1 age_co_driver                               0.26
## 2 confidence                                  0.34
## 3 driving_years_drone                         0.27
## 4 inconsistent_codriver                       0.31
## 5 neuroticism                                 0.31
## 6 switch_cost                                -0.34
## 7 terrible_codriver                           0.5 
##                       vars  n mean sd median trimmed  mad   min  max range
## age_co_driver            1 52    0  1  -0.31   -0.21 0.27 -0.49 6.12  6.61
## confidence               2 52    0  1  -0.01    0.03 0.77 -2.71 1.83  4.55
## driving_years_drone      3 52    0  1  -0.23   -0.19 0.29 -0.43 6.38  6.80
## inconsistent_codriver    4 52    0  1  -0.05   -0.07 1.23 -1.38 2.48  3.85
## neuroticism              5 52    0  1  -0.09    0.03 1.27 -2.15 1.62  3.77
## switch_cost              6 52    0  1  -0.15   -0.05 0.98 -2.30 2.53  4.82
## terrible_codriver        7 52    0  1  -0.16   -0.15 0.73 -1.37 3.55  4.93
##                        skew kurtosis   se
## age_co_driver          4.78    24.89 0.14
## confidence            -0.36     0.35 0.14
## driving_years_drone    5.20    29.32 0.14
## inconsistent_codriver  0.57    -0.53 0.14
## neuroticism           -0.11    -0.98 0.14
## switch_cost            0.37    -0.35 0.14
## terrible_codriver      1.56     2.77 0.14
## 
## Call:
## lm(formula = tmp$distance_overall_deviation ~ ., data = var_std)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -1840.1  -621.5  -144.5   566.3  2990.2 
## 
## Coefficients:
##                       Estimate Std. Error t value Pr(>|t|)    
## (Intercept)              878.1      138.0   6.364 9.84e-08 ***
## age_co_driver           -198.5      467.0  -0.425   0.6729    
## confidence               287.9      151.6   1.899   0.0641 .  
## driving_years_drone      545.7      472.2   1.156   0.2541    
## inconsistent_codriver    131.8      153.1   0.861   0.3940    
## neuroticism              397.3      148.2   2.680   0.0103 *  
## switch_cost             -225.8      158.6  -1.423   0.1617    
## terrible_codriver        730.4      156.2   4.676 2.79e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 995 on 44 degrees of freedom
## Multiple R-squared:  0.5665, Adjusted R-squared:  0.4975 
## F-statistic: 8.215 on 7 and 44 DF,  p-value: 2.239e-06
## 
##                       Correlations                       
## --------------------------------------------------------
## Variable                 Zero Order    Partial     Part  
## --------------------------------------------------------
## age_co_driver                 0.260     -0.064    -0.042 
## confidence                    0.344      0.275     0.188 
## driving_years_drone           0.269      0.172     0.115 
## inconsistent_codriver         0.314      0.129     0.085 
## neuroticism                   0.305      0.375     0.266 
## switch_cost                  -0.336     -0.210    -0.141 
## terrible_codriver             0.496      0.576     0.464 
## --------------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.0585         0
## Farrar Chi-Square:       137.0310         1
## Red Indicator:             0.2517         0
## Sum of Lambda Inverse:    28.7970         0
## Theil's Method:           -0.6986         0
## Condition Number:         36.5009         1
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                           VIF    TOL      Wi      Fi Leamer    CVIF Klein
## age_co_driver         11.2337 0.0890 76.7524 94.1496 0.2984 25.4736     1
## confidence             1.1838 0.8448  1.3782  1.6906 0.9191  2.6843     0
## driving_years_drone   11.4870 0.0871 78.6526 96.4805 0.2951 26.0481     1
## inconsistent_codriver  1.2076 0.8281  1.5570  1.9099 0.9100  2.7384     0
## neuroticism            1.1319 0.8834  0.9896  1.2139 0.9399  2.5668     0
## switch_cost            1.2963 0.7714  2.2225  2.7263 0.8783  2.9396     0
## terrible_codriver      1.2567 0.7957  1.9251  2.3615 0.8920  2.8497     0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## age_co_driver , confidence , driving_years_drone , inconsistent_codriver , switch_cost , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.5665 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [1] "Team 16110214_1 removed from analyses"
## # A tibble: 5 x 2
##   rowname             distance_overall_deviation
##   <chr>                                    <dbl>
## 1 driving_years_drone                       0.27
## 2 incongruent_errors                        0.35
## 3 neuroticism                               0.43
## 4 switch_cost                              -0.4 
## 5 wm_accuracy_drone                        -0.26
##                     vars  n mean sd median trimmed  mad   min  max range
## driving_years_drone    1 52    0  1  -0.23   -0.19 0.29 -0.43 6.37  6.80
## incongruent_errors     2 52    0  1  -0.33   -0.16 0.83 -0.89 3.03  3.92
## neuroticism            3 52    0  1   0.04    0.03 1.24 -2.14 1.55  3.69
## switch_cost            4 52    0  1  -0.34   -0.06 1.01 -2.27 2.54  4.82
## wm_accuracy_drone      5 52    0  1  -0.10   -0.04 0.86 -2.24 2.42  4.67
##                      skew kurtosis   se
## driving_years_drone  5.20    29.29 0.14
## incongruent_errors   1.17     0.54 0.14
## neuroticism         -0.14    -1.02 0.14
## switch_cost          0.42    -0.33 0.14
## wm_accuracy_drone    0.35    -0.33 0.14
## 
## Call:
## lm(formula = tmp$distance_overall_deviation ~ ., data = var_std)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2304.3  -888.5    37.7   597.0  3258.0 
## 
## Coefficients:
##                     Estimate Std. Error t value Pr(>|t|)    
## (Intercept)            876.3      163.9   5.345 2.74e-06 ***
## driving_years_drone    141.1      190.6   0.740   0.4629    
## incongruent_errors     188.7      199.1   0.948   0.3483    
## neuroticism            440.6      174.8   2.521   0.0152 *  
## switch_cost           -325.8      184.4  -1.767   0.0839 .  
## wm_accuracy_drone     -280.7      171.5  -1.636   0.1086    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1182 on 46 degrees of freedom
## Multiple R-squared:  0.3547, Adjusted R-squared:  0.2846 
## F-statistic: 5.057 on 5 and 46 DF,  p-value: 0.0008929
## 
##                      Correlations                      
## ------------------------------------------------------
## Variable               Zero Order    Partial     Part  
## ------------------------------------------------------
## driving_years_drone         0.267      0.109     0.088 
## incongruent_errors          0.355      0.138     0.112 
## neuroticism                 0.427      0.348     0.299 
## switch_cost                -0.400     -0.252    -0.209 
## wm_accuracy_drone          -0.258     -0.235    -0.194 
## ------------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.5771         0
## Farrar Chi-Square:        26.3906         1
## Red Indicator:             0.2389         0
## Sum of Lambda Inverse:     6.2015         0
## Theil's Method:           -0.4988         0
## Condition Number:         13.1878         0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                        VIF    TOL     Wi     Fi Leamer   CVIF Klein
## driving_years_drone 1.3252 0.7546 3.8211 5.2033 0.8687 2.1736     0
## incongruent_errors  1.4468 0.6912 5.2505 7.1496 0.8314 2.3732     0
## neuroticism         1.1152 0.8967 1.3538 1.8435 0.9469 1.8292     0
## switch_cost         1.2409 0.8059 2.8307 3.8546 0.8977 2.0354     0
## wm_accuracy_drone   1.0734 0.9317 0.8620 1.1738 0.9652 1.7606     0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## driving_years_drone , incongruent_errors , switch_cost , wm_accuracy_drone , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.3547 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [1] "Team 17032409_1 removed from analyses"
## # A tibble: 8 x 2
##   rowname               distance_overall_deviation
##   <chr>                                      <dbl>
## 1 confidence                                  0.3 
## 2 congruent_time                             -0.26
## 3 driving_years_drone                         0.26
## 4 incongruent_errors                          0.34
## 5 inconsistent_codriver                       0.28
## 6 neuroticism                                 0.33
## 7 switch_cost                                -0.26
## 8 terrible_codriver                           0.42
##                       vars  n mean sd median trimmed  mad   min  max range
## confidence               1 52    0  1   0.00    0.03 0.80 -2.69 1.83  4.52
## congruent_time           2 52    0  1  -0.17   -0.08 0.89 -1.61 2.81  4.42
## driving_years_drone      3 52    0  1  -0.23   -0.19 0.29 -0.43 6.37  6.80
## incongruent_errors       4 52    0  1  -0.33   -0.16 0.83 -0.89 3.03  3.92
## inconsistent_codriver    5 52    0  1  -0.04   -0.07 1.23 -1.37 2.47  3.84
## neuroticism              6 52    0  1  -0.10    0.02 1.26 -2.13 1.60  3.73
## switch_cost              7 52    0  1  -0.20   -0.08 0.98 -1.39 2.63  4.02
## terrible_codriver        8 52    0  1  -0.17   -0.15 0.75 -1.34 3.55  4.89
##                        skew kurtosis   se
## confidence            -0.34     0.27 0.14
## congruent_time         0.78     0.00 0.14
## driving_years_drone    5.20    29.29 0.14
## incongruent_errors     1.17     0.54 0.14
## inconsistent_codriver  0.58    -0.55 0.14
## neuroticism           -0.09    -0.97 0.14
## switch_cost            0.60    -0.48 0.14
## terrible_codriver      1.57     2.72 0.14
## 
## Call:
## lm(formula = tmp$distance_overall_deviation ~ ., data = var_std)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2044.0  -646.8  -111.1   529.2  4065.0 
## 
## Coefficients:
##                       Estimate Std. Error t value Pr(>|t|)    
## (Intercept)             891.78     156.83   5.686 1.04e-06 ***
## confidence              340.86     219.22   1.555 0.127315    
## congruent_time           20.26     251.62   0.081 0.936191    
## driving_years_drone     215.40     182.30   1.182 0.243860    
## incongruent_errors      286.17     228.52   1.252 0.217245    
## inconsistent_codriver   129.20     173.81   0.743 0.461332    
## neuroticism             474.37     171.01   2.774 0.008160 ** 
## switch_cost             -12.42     201.24  -0.062 0.951081    
## terrible_codriver       610.40     172.21   3.544 0.000963 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1131 on 43 degrees of freedom
## Multiple R-squared:  0.483,  Adjusted R-squared:  0.3868 
## F-statistic: 5.021 on 8 and 43 DF,  p-value: 0.0001947
## 
##                       Correlations                       
## --------------------------------------------------------
## Variable                 Zero Order    Partial     Part  
## --------------------------------------------------------
## confidence                    0.299      0.231     0.170 
## congruent_time               -0.260      0.012     0.009 
## driving_years_drone           0.255      0.177     0.130 
## incongruent_errors            0.339      0.188     0.137 
## inconsistent_codriver         0.280      0.113     0.082 
## neuroticism                   0.327      0.390     0.304 
## switch_cost                  -0.256     -0.009    -0.007 
## terrible_codriver             0.417      0.476     0.389 
## --------------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.1615         0
## Farrar Chi-Square:        88.2044         1
## Red Indicator:             0.2489         0
## Sum of Lambda Inverse:    13.0175         0
## Theil's Method:           -0.6859         0
## Condition Number:         49.8746         1
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                          VIF    TOL     Wi      Fi Leamer   CVIF Klein
## confidence            1.9165 0.5218 5.7608  6.8737 0.7223 4.1601     0
## congruent_time        2.5248 0.3961 9.5845 11.4360 0.6293 5.4806     1
## driving_years_drone   1.3252 0.7546 2.0443  2.4393 0.8687 2.8767     0
## incongruent_errors    2.0825 0.4802 6.8042  8.1187 0.6930 4.5204     1
## inconsistent_codriver 1.2047 0.8301 1.2867  1.5353 0.9111 2.6150     0
## neuroticism           1.1662 0.8575 1.0448  1.2467 0.9260 2.5315     0
## switch_cost           1.6149 0.6192 3.8652  4.6118 0.7869 3.5055     0
## terrible_codriver     1.1827 0.8456 1.1481  1.3699 0.9195 2.5672     0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## confidence , congruent_time , driving_years_drone , incongruent_errors , inconsistent_codriver , switch_cost , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.483 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [1] "Team 17081512_2 removed from analyses"
## # A tibble: 5 x 2
##   rowname            distance_overall_deviation
##   <chr>                                   <dbl>
## 1 confidence                               0.26
## 2 incongruent_errors                       0.3 
## 3 neuroticism                              0.37
## 4 switch_cost                             -0.34
## 5 terrible_codriver                        0.4 
##                    vars  n mean sd median trimmed  mad   min  max range
## confidence            1 52    0  1  -0.03    0.02 0.78 -2.67 1.84  4.51
## incongruent_errors    2 52    0  1  -0.32   -0.16 0.83 -0.88 3.04  3.92
## neuroticism           3 52    0  1  -0.10    0.02 1.26 -2.13 1.60  3.73
## switch_cost           4 52    0  1  -0.17   -0.05 0.94 -2.33 2.54  4.86
## terrible_codriver     5 52    0  1  -0.15   -0.15 0.74 -1.34 3.63  4.97
##                     skew kurtosis   se
## confidence         -0.31     0.24 0.14
## incongruent_errors  1.20     0.58 0.14
## neuroticism        -0.09    -0.97 0.14
## switch_cost         0.37    -0.30 0.14
## terrible_codriver   1.67     3.20 0.14
## 
## Call:
## lm(formula = tmp$distance_overall_deviation ~ ., data = var_std)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2372.8  -630.8  -122.1   629.1  3584.5 
## 
## Coefficients:
##                    Estimate Std. Error t value Pr(>|t|)    
## (Intercept)           962.4      163.1   5.899  4.1e-07 ***
## confidence            283.5      177.3   1.599 0.116772    
## incongruent_errors    271.2      179.8   1.508 0.138327    
## neuroticism           564.5      176.5   3.199 0.002498 ** 
## switch_cost          -227.2      194.8  -1.166 0.249530    
## terrible_codriver     700.5      168.1   4.168 0.000134 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1176 on 46 degrees of freedom
## Multiple R-squared:  0.4709, Adjusted R-squared:  0.4134 
## F-statistic: 8.187 on 5 and 46 DF,  p-value: 1.364e-05
## 
##                     Correlations                      
## -----------------------------------------------------
## Variable              Zero Order    Partial     Part  
## -----------------------------------------------------
## confidence                 0.260      0.229     0.171 
## incongruent_errors         0.303      0.217     0.162 
## neuroticism                0.366      0.427     0.343 
## switch_cost               -0.341     -0.169    -0.125 
## terrible_codriver          0.404      0.524     0.447 
## -----------------------------------------------------

## 
## Call:
## omcdiag(x = x, y = y)
## 
## 
## Overall Multicollinearity Diagnostics
## 
##                        MC Results detection
## Determinant |X'X|:         0.6548         0
## Farrar Chi-Square:        20.3218         1
## Red Indicator:             0.2017         0
## Sum of Lambda Inverse:     5.9389         0
## Theil's Method:           -1.1319         0
## Condition Number:         14.7394         0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## 
## Call:
## imcdiag(x = x, y = y)
## 
## 
## All Individual Multicollinearity Diagnostics Result
## 
##                       VIF    TOL     Wi     Fi Leamer   CVIF Klein
## confidence         1.1589 0.8629 1.8666 2.5418 0.9289 1.4362     0
## incongruent_errors 1.1920 0.8389 2.2563 3.0724 0.9159 1.4773     0
## neuroticism        1.1477 0.8713 1.7357 2.3635 0.9334 1.4224     0
## switch_cost        1.3993 0.7147 4.6914 6.3883 0.8454 1.7341     0
## terrible_codriver  1.0410 0.9606 0.4816 0.6558 0.9801 1.2901     0
## 
## 1 --> COLLINEARITY is detected by the test 
## 0 --> COLLINEARITY is not detected by the test
## 
## confidence , incongruent_errors , switch_cost , coefficient(s) are non-significant may be due to multicollinearity
## 
## R-square of y on all x: 0.4709 
## 
## * use method argument to check which regressors may be the reason of collinearity
## ===================================
## [[1]]
## NULL
## 
## [[2]]
## NULL
## 
## [[3]]
## NULL
## 
## [[4]]
## NULL

Investigate distance deviation outliers

Not sure why the regression results change so much when any of these outliers are removed. Need to investigate further.

When did the deviation occur for each of the outlier teams?

All teams had good network connections. The codrivers all saw the traffic experienced by the drivers and the arrows appeared correctly on the codrivers screens.

For teams 16110214_1 and 17081510_2 a huge deviation occurred during lap 4. I have checked the screen capture videos for these 2 teams.

For team 16110214_1 the codriver paid no attention to the arrows throughout the whole drive. They were often saying “keep going straight” regardless of the direction the arrows pointed. During lap 4 they directed the driver the wrong way twice at the same black ice intersection. This sent the driver on a large loop. The first time they told the driver to turn right (instead of left), the second time they directed the driver straight and the third time the driver was directed left. The driver listened to the codriver’s direction even when they could see it was wrong. For substantial periods of lap 4 the codriver was not following the driver with the drone but they continued to direct and offer info/instruction. The codriver clearly did not follow instructions. This team should be high on inconsistent_codriver and maybe terrible codriver.

For team 17081510_2 the misdirection occurred at the same black ice intersection in lap 4 although this time it was a design flaw that sent the driver the wrong way. They entered the intersection with the intention of turning right but when they lost control on the black ice and overshot the turn they could see green lights in the distance for the next intersection and they were pointing straight so they continued in that direction. This happened exactly the same a second time when they overshot the black ice intersection. On the third attempt to turn left they did not overshoot it and made it. It seems the driver was following instructions but a design flaw sent them the wrong way. This should not influence the teams scores on the comms factors.

For team 16101114_1 large deviations occured in laps 1 and 5. During lap 1 the driver was not doing a good job of monitoring the arrows and made a number of wrong turns that went uncorrected because the codriver was also not paying attention to the arrows. During lap 5 the driver missed a turn (again they didn’t notice) and when the codriver tried to correct them the driver could see arrows at the next intersection and decided to continue in the wrong direction. This team should score positively on helpful codriver.

Let’s check the comms factor scores for each of these teams.

## # A tibble: 3 x 4
##   team       inconsistent_codriver terrible_codriver helpful_exchange
##   <chr>                      <dbl>             <dbl>            <dbl>
## 1 16101114_1                -0.656            -0.925            0.855
## 2 16110214_1                 0.351             3.60            -0.795
## 3 17032409_1                -0.297             0.126            1.11

I recoded the comms for 16110214_1. How do the new scores on the comms variables and comms factors compare to the original scores.

## [1] "New scores"
## # A tibble: 16 x 2
##    var                          val
##    <chr>                      <dbl>
##  1 inconsistent_codriver      0.351
##  2 terrible_codriver          3.60 
##  3 helpful_exchange          -0.795
##  4 co_info_help_overall      21    
##  5 co_info_harm_overall      14    
##  6 co_instruct_help_overall  70    
##  7 co_instruct_harm_overall  22    
##  8 co_total_help_overall     91    
##  9 co_total_harm_overall     36    
## 10 co_redundant_overall      33    
## 11 co_question_overall        3    
## 12 co_total_overall         163    
## 13 drive_question_overall    42    
## 14 drive_informs_overall     31    
## 15 drive_frust_overall       35    
## 16 drive_total_overall       73
## [1] "Original scores"
## # A tibble: 16 x 2
##    var                         val
##    <chr>                     <dbl>
##  1 inconsistent_codriver      1.33
##  2 terrible_codriver          2.37
##  3 helpful_codriver          -1.47
##  4 co_info_help_overall      20   
##  5 co_info_harm_overall      12   
##  6 co_instruct_help_overall  79   
##  7 co_instruct_harm_overall  29   
##  8 co_total_help_overall     99   
##  9 co_total_harm_overall     41   
## 10 co_redundant_overall      17   
## 11 co_question_overall        3   
## 12 co_total_overall         160   
## 13 drive_question_overall    40   
## 14 drive_informs_overall     32   
## 15 drive_frust_overall       29   
## 16 drive_total_overall       72

For team 17081512_2 there was no large deviation so they were not an outlier on the DV (distance deviation). Maybe they were an outlier on the comms factors. Let’s take a look.

17081512_2 were initially identified as having a large influence on the distance deviation results because they had a very high score on the terrible codriver factor. I reviewed the coding and made some changes. How do the new scores on the comms variables and comms factors compare to the original scores.

## [1] "New scores"
## # A tibble: 16 x 2
##    var                          val
##    <chr>                      <dbl>
##  1 inconsistent_codriver     -0.820
##  2 terrible_codriver          1.30 
##  3 helpful_exchange           0.868
##  4 co_info_help_overall      81    
##  5 co_info_harm_overall       7    
##  6 co_instruct_help_overall  48    
##  7 co_instruct_harm_overall   3    
##  8 co_total_help_overall    129    
##  9 co_total_harm_overall     10    
## 10 co_redundant_overall      27    
## 11 co_question_overall        1    
## 12 co_total_overall         167    
## 13 drive_question_overall    32    
## 14 drive_informs_overall     26    
## 15 drive_frust_overall       30    
## 16 drive_total_overall       58
## [1] "Original scores"
## # A tibble: 16 x 2
##    var                          val
##    <chr>                      <dbl>
##  1 inconsistent_codriver     -0.991
##  2 terrible_codriver          4.10 
##  3 helpful_codriver           1.09 
##  4 co_info_help_overall      78    
##  5 co_info_harm_overall      22    
##  6 co_instruct_help_overall  70    
##  7 co_instruct_harm_overall   2    
##  8 co_total_help_overall    148    
##  9 co_total_harm_overall     24    
## 10 co_redundant_overall      51    
## 11 co_question_overall        1    
## 12 co_total_overall         224    
## 13 drive_question_overall    40    
## 14 drive_informs_overall     33    
## 15 drive_frust_overall       33    
## 16 drive_total_overall       73

Investigate speed outliers

I recoded the comms for each of the speed outliers because the results were different with each outlier removed separately. How do the new scores on the comms variables and comms factors compare to the original scores.

## [1] "New scores"
## # A tibble: 3 x 4
##   team       inconsistent_codriver terrible_codriver helpful_exchange
##   <chr>                      <dbl>             <dbl>            <dbl>
## 1 17040414_2               -1.17              -1.37            -1.11 
## 2 17040711_1               -0.414             -0.914            0.297
## 3 17080810_2                0.0783             0.348           -0.872
## [1] "Original scores"
## # A tibble: 3 x 4
##   team       inconsistent_codriver terrible_codriver helpful_codriver
##   <chr>                      <dbl>             <dbl>            <dbl>
## 1 17040414_2                -1.21             -1.20            -1.10 
## 2 17040711_1                -0.151            -0.672           -0.426
## 3 17080810_2                 0.495             0.481           -0.973